Research And Development Of Water-fertilizer Integrated Sprinkler Irrigation Device Based On The Principle Of Runoff Prevention In Dry Depth Time Domai

Water-fertilizer integrated sprinkler irrigation equipment has many advantages such as water saving,energy saving,etc.It plays a pivotal role in China’s goal of modernizing agriculture and becoming carbon neutral.Water-fertilizer integrated sprinkler irrigation is one of the more commonly used irrigation technologies in the world,which can create a suitable nutrient environment for crops in the field and promote their growth and development.At the same time,water-fertilizer integrated sprinkler irrigation equipment is highly adaptable and can be used in different terrains and soil environments for irrigation.However,when water-fertilizer integrated sprinklers are not designed properly,it will lead to soil runoff,which will cause nutrients to be washed away from the soil surface,reducing the productivity of the soil and wasting water resources while polluting the river water.At the same time,soil runoff also makes the soil dry and the water storage capacity in the soil layer decreases,which is not conducive to the normal growth of crops.Therefore,it is of great practical value to study water-fertilizer integrated runoff prevention sprinkler irrigation equipment.In this thesis,in order to improve the efficiency of water-fertilizer integrated sprinkler irrigation equipment and at the same time avoid runoff from the soil surface,the theory of unsaturated soil infiltration,neural network and finite element ideas are combined with sprinklers to develop water-fertilizer integrated sprinklers that prevent runoff from the surface.The main research of this paper is as follows:1)To address the problem that the radial base center parameters of the traditional RBF neural network are difficult to determine,resulting in the poor effect of the neural network,this paper proposes to use the MOV algorithm to cluster to obtain the radial base center values of the RBF neural network,and then use the least squares method to obtain the weighting and bias values of the RBF neural network.2)In order to obtain the application rate of sprinkler irrigation of runoff prevention,a sprinkler irrigation soil infiltration model is established by unsaturated soil infiltration theory and crop retention theory,and the application rate of sprinkler irrigation corresponding to 1376 groups of different soil textures,initial water content,soil bulk density and effective irrigation volume are simulated.The data obtained from the simulation were then divided into training and test sets,and the MOV-RBF neural network prediction model was established with effective irrigation volume,soil bulk density,initial water content and soil texture as inputs,and it was compared with other neural networks and the accuracy of MOV-RBF neural network was obtained to be the highest among them.3)In order to realize the runoff prevention function of water and fertilizer integrated sprinkler,the neural network runoff prevention application rate of sprinkler irrigation prediction model,empirical formula and PID control are combined with sprinkler to develop water and fertilizer integrated runoff prevention sprinkler.4)In order to verify the accuracy of the neural network prediction model in practice and the superiority of the integrated fertilizer sprinkler,the experimental runoff prevention application rate of sprinkler irrigation is compared with the prediction result of MOV-RBF neural network to obtain the accuracy of the prediction model in practice.Then,this paper compares the water and fertilizer integrated runoff prevention sprinkler with other water and fertilizer integrated sprinklers to obtain the effectiveness and superiority of the water and fertilizer integrated runoff prevention sprinkler developed in this paper in practice.